No abstract

The selectivity of a complex reaction depends on a number of factors, such as the reaction mechanism, operating conditions, catalyst properties and catalyst deactivation. The present work discusses how the selectivity of a complex reaction depends on the formation of coke. For zeolite catalysts, changes in selectivity can be a result of intrinsic selectivity effects or shape selectivity effects. A method is suggested to analyze a complex reaction system with deactivation caused by coke formation, and different cases of selectivity change during deactivation of zeolites are discussed. Transition-state shape selective deactivation is proposed as a mechanism in addition to the deactivation mechanisms suggested by Guisnet and Magnoux (Guisnet, M.; Magnoux, P. Appl. Catal. 1989, 54, 1). By variation of the space velocity, the selectivities to the main products are measured as a function of conversion (optimum performance envelopes). Selectivities at given conversions can then be compared for results obtained with different contact time and with varying degree of catalyst deactivation (space velocity loops). Selective or nonselective deactivation is thereby distinguished. This type of selectivity plot is applied to two different types of reaction, i.e. ethene oligomerization over HZSM-5 and methanol conversion to light olefins (MTO) over SAPO-34. The selectivity of ethene oligomerization was affected only by the decrease in conversion due to coke formation; hence, this is an example of nonselective deactivation. Selective deactivation was found for methanol conversion over SAPO-34.

The effects of adsorption and diffusion of the reactants on methanol to olefins (MTO) and propene conversion over SAPO-34 have been studied in an oscillating microbalance reactor. The adsorption parameters of methanol and propene at reaction conditions (698 K) were determined by a pulse method, and the results were identical to the values obtained by extrapolation from low temperatures (323−398 K). Inverse uptake diffusion times were calculated from adsorption data at low temperatures, and these results were dependent on the temperature and the adsorbed amount. The inverse steady-state diffusion times calculated from the inverse uptake diffusion times were independent of the temperature and the adsorbed amount. The influence of diffusion on the reaction rates was estimated on the basis of the inverse steady-state diffusion time, using the Weisz−Prater criterion. The methanol conversion over SAPO-34 was influenced by diffusion of the reactant, while the propene conversion was not. A kinetic study revealed that both the rate constant and the site coverage of propene were much lower than that of methanol at 698 K. The deactivation behavior during the MTO reaction over SAPO-34 was studied by measuring both the adsorbed amount of methanol and the conversion at different coke contents. Catalyst deactivation was proposed to be due to a decreasing number of sites available for adsorption at high coke contents and a lower diffusivity, hence a lower effectiveness factor due to coke deposition.